Building Extraction from High-Resolution Aerial Imagery Using a Generative Adversarial Network with Spatial and Channel Attention Mechanisms
Segmentation of high-resolution remote sensing images is an important challenge with wide practical applications. The increasing spatial resolution provides fine details for image segmentation but also incurs segmentation ambiguities. In this paper, we propose a generative adversarial network with s...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-04-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/11/8/917 |
_version_ | 1798030953703014400 |
---|---|
author | Xuran Pan Fan Yang Lianru Gao Zhengchao Chen Bing Zhang Hairui Fan Jinchang Ren |
author_facet | Xuran Pan Fan Yang Lianru Gao Zhengchao Chen Bing Zhang Hairui Fan Jinchang Ren |
author_sort | Xuran Pan |
collection | DOAJ |
description | Segmentation of high-resolution remote sensing images is an important challenge with wide practical applications. The increasing spatial resolution provides fine details for image segmentation but also incurs segmentation ambiguities. In this paper, we propose a generative adversarial network with spatial and channel attention mechanisms (GAN-SCA) for the robust segmentation of buildings in remote sensing images. The segmentation network (generator) of the proposed framework is composed of the well-known semantic segmentation architecture (U-Net) and the spatial and channel attention mechanisms (SCA). The adoption of SCA enables the segmentation network to selectively enhance more useful features in specific positions and channels and enables improved results closer to the ground truth. The discriminator is an adversarial network with channel attention mechanisms that can properly discriminate the outputs of the generator and the ground truth maps. The segmentation network and adversarial network are trained in an alternating fashion on the Inria aerial image labeling dataset and Massachusetts buildings dataset. Experimental results show that the proposed GAN-SCA achieves a higher score (the overall accuracy and intersection over the union of Inria aerial image labeling dataset are 96.61% and 77.75%, respectively, and the F<sub>1</sub>-measure of the Massachusetts buildings dataset is 96.36%) and outperforms several state-of-the-art approaches. |
first_indexed | 2024-04-11T19:49:35Z |
format | Article |
id | doaj.art-c0e3f749da2c4975afbe9ab6f02ca32d |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-04-11T19:49:35Z |
publishDate | 2019-04-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-c0e3f749da2c4975afbe9ab6f02ca32d2022-12-22T04:06:21ZengMDPI AGRemote Sensing2072-42922019-04-0111891710.3390/rs11080917rs11080917Building Extraction from High-Resolution Aerial Imagery Using a Generative Adversarial Network with Spatial and Channel Attention MechanismsXuran Pan0Fan Yang1Lianru Gao2Zhengchao Chen3Bing Zhang4Hairui Fan5Jinchang Ren6School of Electronics and Information Engineering, Hebei University of Technology, Tianjin 300401, ChinaSchool of Electronics and Information Engineering, Hebei University of Technology, Tianjin 300401, ChinaKey Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, ChinaSchool of Electronics and Information Engineering, Hebei University of Technology, Tianjin 300401, ChinaDepartment of Electronic and Electrical Engineering, University of Strathclyde, Glasgow, G1 1XW, UKSegmentation of high-resolution remote sensing images is an important challenge with wide practical applications. The increasing spatial resolution provides fine details for image segmentation but also incurs segmentation ambiguities. In this paper, we propose a generative adversarial network with spatial and channel attention mechanisms (GAN-SCA) for the robust segmentation of buildings in remote sensing images. The segmentation network (generator) of the proposed framework is composed of the well-known semantic segmentation architecture (U-Net) and the spatial and channel attention mechanisms (SCA). The adoption of SCA enables the segmentation network to selectively enhance more useful features in specific positions and channels and enables improved results closer to the ground truth. The discriminator is an adversarial network with channel attention mechanisms that can properly discriminate the outputs of the generator and the ground truth maps. The segmentation network and adversarial network are trained in an alternating fashion on the Inria aerial image labeling dataset and Massachusetts buildings dataset. Experimental results show that the proposed GAN-SCA achieves a higher score (the overall accuracy and intersection over the union of Inria aerial image labeling dataset are 96.61% and 77.75%, respectively, and the F<sub>1</sub>-measure of the Massachusetts buildings dataset is 96.36%) and outperforms several state-of-the-art approaches.https://www.mdpi.com/2072-4292/11/8/917high-resolution aerial imagesdeep learninggenerative adversarial networksemantic segmentationInria aerial image labeling datasetMassachusetts buildings dataset |
spellingShingle | Xuran Pan Fan Yang Lianru Gao Zhengchao Chen Bing Zhang Hairui Fan Jinchang Ren Building Extraction from High-Resolution Aerial Imagery Using a Generative Adversarial Network with Spatial and Channel Attention Mechanisms Remote Sensing high-resolution aerial images deep learning generative adversarial network semantic segmentation Inria aerial image labeling dataset Massachusetts buildings dataset |
title | Building Extraction from High-Resolution Aerial Imagery Using a Generative Adversarial Network with Spatial and Channel Attention Mechanisms |
title_full | Building Extraction from High-Resolution Aerial Imagery Using a Generative Adversarial Network with Spatial and Channel Attention Mechanisms |
title_fullStr | Building Extraction from High-Resolution Aerial Imagery Using a Generative Adversarial Network with Spatial and Channel Attention Mechanisms |
title_full_unstemmed | Building Extraction from High-Resolution Aerial Imagery Using a Generative Adversarial Network with Spatial and Channel Attention Mechanisms |
title_short | Building Extraction from High-Resolution Aerial Imagery Using a Generative Adversarial Network with Spatial and Channel Attention Mechanisms |
title_sort | building extraction from high resolution aerial imagery using a generative adversarial network with spatial and channel attention mechanisms |
topic | high-resolution aerial images deep learning generative adversarial network semantic segmentation Inria aerial image labeling dataset Massachusetts buildings dataset |
url | https://www.mdpi.com/2072-4292/11/8/917 |
work_keys_str_mv | AT xuranpan buildingextractionfromhighresolutionaerialimageryusingagenerativeadversarialnetworkwithspatialandchannelattentionmechanisms AT fanyang buildingextractionfromhighresolutionaerialimageryusingagenerativeadversarialnetworkwithspatialandchannelattentionmechanisms AT lianrugao buildingextractionfromhighresolutionaerialimageryusingagenerativeadversarialnetworkwithspatialandchannelattentionmechanisms AT zhengchaochen buildingextractionfromhighresolutionaerialimageryusingagenerativeadversarialnetworkwithspatialandchannelattentionmechanisms AT bingzhang buildingextractionfromhighresolutionaerialimageryusingagenerativeadversarialnetworkwithspatialandchannelattentionmechanisms AT hairuifan buildingextractionfromhighresolutionaerialimageryusingagenerativeadversarialnetworkwithspatialandchannelattentionmechanisms AT jinchangren buildingextractionfromhighresolutionaerialimageryusingagenerativeadversarialnetworkwithspatialandchannelattentionmechanisms |